scour/scour.py

#!/usr/bin/env python

#  Scour
#  Version 0.08
#
#  Copyright 2009 Jeff Schiller
#
#  This file is part of Scour, http://www.codedread.com/scour/
#
#   Licensed under the Apache License, Version 2.0 (the "License");
#   you may not use this file except in compliance with the License.
#   You may obtain a copy of the License at
#
#       http://www.apache.org/licenses/LICENSE-2.0
#
#   Unless required by applicable law or agreed to in writing, software
#   distributed under the License is distributed on an "AS IS" BASIS,
#   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#   See the License for the specific language governing permissions and
#   limitations under the License.

# Notes:

# rubys' path-crunching ideas here: http://intertwingly.net/code/svgtidy/spec.rb
# (and implemented here: http://intertwingly.net/code/svgtidy/svgtidy.rb )

# Yet more ideas here: http://wiki.inkscape.org/wiki/index.php/Save_Cleaned_SVG
# TODO: Adapt this script into an Inkscape python plugin
#
# * Specify a limit to the precision of all positional elements.
# * Clean up Definitions
#  * Collapse duplicate gradient definitions
#  * Remove gradients that are only referenced by one other gradient
# * Clean up CSS
#  * Convert RGB colours from RGB(r,g,b) to #RRGGBB format
#  * Convert RGB colours from #RRGGBB to #RGB if possible
# * Clean up paths
#  * Detect vertical/horizontal lines and replace.
#  * Eliminate empty path segments
#  * Eliminate last segment in a polygon
#  * Collapse straight curves.
#  * Convert absolute path segments to relative ones.
# * Process Transformations
#  * Process quadratic Bezier curves
#  * Collapse all group based transformations

# Next Up:
# + Remove unnecessary nested <g> elements
# + Remove duplicate gradient stops (same offset, stop-color, stop-opacity)
# + Always keep fonts inside <defs>
# - Convert all colors to #RRGGBB format
# - Reduce #RRGGBB format to #RGB format when possible
# - rework command-line argument processing so that options are configurable
# https://bugs.edge.launchpad.net/ubuntu/+source/human-icon-theme/+bug/361667/

# Some notes to not forget:
# - removing unreferenced IDs loses some semantic information
# - removing empty nested groups also potentially loses some semantic information
#   (i.e. the following button:
#     <g>
#       <rect .../>
#       <text .../>
#     </g>
#   will be flattened)


# necessary to get true division
from __future__ import division

import os
import sys
import string
import xml.dom.minidom
import re
import math
import base64
import os.path
import urllib

APP = 'scour'
VER = '0.08'
COPYRIGHT = 'Copyright Jeff Schiller, 2009'

NS = { 	'SVG': 		'http://www.w3.org/2000/svg', 
		'XLINK': 	'http://www.w3.org/1999/xlink', 
		'SODIPODI': 'http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd',
		'INKSCAPE': 'http://www.inkscape.org/namespaces/inkscape',
		'ADOBE_ILLUSTRATOR': 'http://ns.adobe.com/AdobeIllustrator/10.0/',
		'ADOBE_GRAPHS': 'http://ns.adobe.com/Graphs/1.0/',
		'ADOBE_SVG_VIEWER': 'http://ns.adobe.com/AdobeSVGViewerExtensions/3.0/',
		'ADOBE_VARIABLES': 'http://ns.adobe.com/Variables/1.0/',
		'ADOBE_SFW': 'http://ns.adobe.com/SaveForWeb/1.0/',
		'ADOBE_EXTENSIBILITY': 'http://ns.adobe.com/Extensibility/1.0/',
     	'ADOBE_FLOWS': 'http://ns.adobe.com/Flows/1.0/',
     	'ADOBE_IMAGE_REPLACEMENT': 'http://ns.adobe.com/ImageReplacement/1.0/',     
     	'ADOBE_CUSTOM': 'http://ns.adobe.com/GenericCustomNamespace/1.0/',
     	'ADOBE_XPATH': 'http://ns.adobe.com/XPath/1.0/'
		}

unwanted_ns = [ NS['SODIPODI'], NS['INKSCAPE'], NS['ADOBE_ILLUSTRATOR'],
				NS['ADOBE_GRAPHS'], NS['ADOBE_SVG_VIEWER'], NS['ADOBE_VARIABLES'],
				NS['ADOBE_SFW'], NS['ADOBE_EXTENSIBILITY'], NS['ADOBE_FLOWS'],
				NS['ADOBE_IMAGE_REPLACEMENT'], NS['ADOBE_CUSTOM'], NS['ADOBE_XPATH'] ] 

svgAttributes = [
				'clip-rule',
				'fill',
				'fill-opacity',
				'fill-rule',
				'filter',
				'font-family',
				'font-size',
				'font-stretch',
				'font-style',
				'font-variant',
				'font-weight',
				'line-height',
				'opacity',
				'stop-color',
				'stop-opacity',
				'stroke',
				'stroke-dashoffset',
				'stroke-linecap',
				'stroke-linejoin',
				'stroke-miterlimit',
				'stroke-opacity',
				'stroke-width',
				]

def printHeader():
	print APP , VER
	print COPYRIGHT

def printSyntaxAndQuit():
	printHeader()
	print 'usage: scour.py [-i input.svg] [-o output.svg]\n'
	print 'If the input file is not specified, stdin is used.'
	print 'If the output file is not specified, stdout is used.'
	quit()	

# returns all elements with id attributes
def findElementsWithId(node,elems={}):
	id = node.getAttribute('id')
	if id != '' :
		elems[id] = node
	if node.hasChildNodes() :
		for child in node.childNodes:
			# from http://www.w3.org/TR/DOM-Level-2-Core/idl-definitions.html
			# we are only really interested in nodes of type Element (1)
			if child.nodeType == 1 :
				findElementsWithId(child, elems)
	return elems

# returns the number of times an id is referenced
# currently looks at fill, stroke and xlink:href attributes
def findReferencedElements(node,ids={}):
	# TODO: error here (ids is not cleared upon next invocation), the
	# input argument ids is clunky here (see below how it is called)
	href = node.getAttributeNS(NS['XLINK'],'href')
	
	# if xlink:href is set, then grab the id
	if href != '' and len(href) > 1 and href[0] == '#':
		# we remove the hash mark from the beginning of the id
		id = href[1:]
		if ids.has_key(id) :
			ids[id] += 1
		else:
			ids[id] = 1

	# now get all style properties and the fill, stroke, filter attributes
	styles = string.split(node.getAttribute('style'),';')
	referencingProps = ['fill', 'stroke', 'filter', 'clip-path', 'mask',  'marker-start', 
						'marker-end', 'marker-mid']
	for attr in referencingProps:
		styles.append( string.join([attr,node.getAttribute(attr)],':') )
			
	for style in styles:
		propval = string.split(style,':')
		if len(propval) == 2 :
			prop = propval[0].strip()
			val = propval[1].strip()
			if prop in referencingProps and val != '' and val[0:5] == 'url(#' :
				id = val[5:val.find(')')]
				if ids.has_key(id) :
					ids[id] += 1
				else:
					ids[id] = 1
					
	if node.hasChildNodes() :
		for child in node.childNodes:
			if child.nodeType == 1 :
				findReferencedElements(child, ids)
	return ids

numIDsRemoved = 0
numElemsRemoved = 0
numAttrsRemoved = 0
numRastersEmbedded = 0

# removes the unreferenced ID attributes
# returns the number of ID attributes removed
def removeUnreferencedIDs(referencedIDs, identifiedElements):
	global numIDsRemoved
	keepTags = ['font']
	num = 0;
	for id in identifiedElements.keys():
		node = identifiedElements[id]
		if referencedIDs.has_key(id) == False and not node.nodeName in keepTags:
			node.removeAttribute('id')
			numIDsRemoved += 1
			num += 1
	return num

# returns the number of unreferenced children removed from defs elements
def vacuumDefs(doc):
	global numElemsRemoved
	keepTags = ['font', 'style', 'metadata' ]
	num = 0
	defs = doc.documentElement.getElementsByTagNameNS(NS['SVG'], 'defs')
	for aDef in defs:
		elemsToRemove = []
		for elem in aDef.childNodes:
			if elem.nodeType == 1 and elem.getAttribute('id') == '' and not elem.nodeName in keepTags:
				elemsToRemove.append(elem)
		for elem in elemsToRemove:
			aDef.removeChild(elem)
			numElemsRemoved += 1
			num += 1
	return num

# returns the number of unreferenced gradients or patterns removed from the document
# (this relies on the ids being removed first)
def removeUnreferencedElements(doc):
	global numElemsRemoved
	num = 0
	for tag in ['pattern', 'linearGradient', 'radialGradient'] :
		elems = doc.documentElement.getElementsByTagNameNS(NS['SVG'], tag)
		elemsToRemove = []
		for elem in elems:
			if elem.getAttribute('id') == '' :
				elemsToRemove.append(elem)
		for elem in elemsToRemove:
			elem.parentNode.removeChild(elem)
			numElemsRemoved += 1
			num += 1
	return num
	
def removeNamespacedAttributes(node, namespaces):
	global numAttrsRemoved
	num = 0
	if node.nodeType == 1 :
		# remove all namespace'd attributes from this element
		attrList = node.attributes
		attrsToRemove = []
		for attrNum in range(attrList.length):
			attr = attrList.item(attrNum)
			if attr != None and attr.namespaceURI in namespaces:
				attrsToRemove.append(attr.nodeName)
		for attrName in attrsToRemove :
			num += 1
			numAttrsRemoved += 1
			node.removeAttribute(attrName)
		
		# now recurse for children
		for child in node.childNodes:
			num += removeNamespacedAttributes(child, namespaces)
	return num
	
def removeNamespacedElements(node, namespaces):
	global numElemsRemoved
	num = 0
	if node.nodeType == 1 :
		# remove all namespace'd child nodes from this element
		childList = node.childNodes
		childrenToRemove = []
		for child in childList:
			if child != None and child.namespaceURI in namespaces:
				childrenToRemove.append(child)
		for child in childrenToRemove :
			num += 1
			numElemsRemoved += 1
			node.removeChild(child)
		
		# now recurse for children
		for child in node.childNodes:
			num += removeNamespacedElements(child, namespaces)
	return num

# this walks further and further down the tree, removing groups
# which do not have any attributes and promoting their children
# up one level
def removeNestedGroups(node):
	global numElemsRemoved
	num = 0
	
	groupsToRemove = []
	for child in node.childNodes:
		if child.nodeName == 'g' and child.namespaceURI == NS['SVG'] and len(child.attributes) == 0:
			groupsToRemove.append(child)

	for g in groupsToRemove:
		while g.childNodes.length > 0:
			g.parentNode.insertBefore(g.firstChild, g)
		g.parentNode.removeChild(g)
		numElemsRemoved += 1
		num += 1

	# now recurse for children
	for child in node.childNodes:
		if child.nodeType == 1:
			num += removeNestedGroups(child)		
	return num

def removeDuplicateGradientStops(doc):
	global numElemsRemoved
	num = 0
	
	for gradType in ['linearGradient', 'radialGradient']:
		for grad in doc.getElementsByTagNameNS(NS['SVG'], gradType):
			stops = {}
			stopsToRemove = []
			for stop in grad.getElementsByTagNameNS(NS['SVG'], 'stop'):
				offset = string.atof(stop.getAttribute('offset'))
				color = stop.getAttribute('stop-color')
				opacity = stop.getAttribute('stop-opacity')
				if stops.has_key(offset) :
					oldStop = stops[offset]
					if oldStop[0] == color and oldStop[1] == opacity:
						stopsToRemove.append(stop)
				stops[offset] = [color, opacity]
				
			for stop in stopsToRemove:
				stop.parentNode.removeChild(stop)
				num += 1
				numElemsRemoved += 1
	
	# linear gradients
	return num

coord = re.compile("\\-?\\d+\\.?\\d*")
scinumber = re.compile("[\\-\\+]?(\\d*\\.?)?\\d+[eE][\\-\\+]?\\d+")
number = re.compile("[\\-\\+]?(\\d*\\.?)?\\d+")
sciExponent = re.compile("[eE]([\\-\\+]?\\d+)")
unit = re.compile("(em|ex|px|pt|pc|cm|mm|in|\\%){1,1}$")

class Unit:
	INVALID = -1
	NONE = 0
	PCT = 1
	PX = 2
	PT = 3
	PC = 4
	EM = 5
	EX = 6
	CM = 7
	MM = 8
	IN = 9
	
	@staticmethod
	def get(str):
		if str == None or str == '': return Unit.NONE
		elif str == '%': return Unit.PCT
		elif str == 'px': return Unit.PX
		elif str == 'pt': return Unit.PT
		elif str == 'pc': return Unit.PC
		elif str == 'em': return Unit.EM
		elif str == 'ex': return Unit.EX
		elif str == 'cm': return Unit.CM
		elif str == 'mm': return Unit.MM
		elif str == 'in': return Unit.IN
		return Unit.INVALID
	
class SVGLength:
	def __init__(self, str):
#		print "Parsing '%s'" % str
		try: # simple unitless and no scientific notation
			self.value = string.atof(str)
			self.units = Unit.NONE
#			print "  Value =", self.value
		except ValueError:
			# we know that the length string has an exponent, a unit, both or is invalid

			# TODO: parse out number, exponent and unit
			unitBegin = 0
			scinum = scinumber.match(str)
			if scinum != None:
				# this will always match, no need to check it
				numMatch = number.match(str)
				expMatch = sciExponent.search(str, numMatch.start(0))
				self.value = string.atof(numMatch.group(0)) * math.pow(10, string.atof(expMatch.group(1)))
				unitBegin = expMatch.end(1)
			else:
				# unit or invalid
				numMatch = number.match(str)
				if numMatch != None:
					self.value = string.atof(numMatch.group(0))
					unitBegin = numMatch.end(0)

			if unitBegin != 0 :
#				print "  Value =", self.value
				unitMatch = unit.search(str, unitBegin)
				if unitMatch != None :
					self.units = Unit.get(unitMatch.group(0))
#					print "  Units =", self.units
				
			# invalid
			else:
				# TODO: this needs to set the default for the given attribute (how?)
#				print "  Invalid: ", str
				self.value = 0 
				self.units = Unit.INVALID

# returns the length of a property
# TODO: eventually use the above class once it is complete
def getSVGLength(value):
	try:
		v = string.atof(value)
	except ValueError:
		coordMatch = coord.match(value)
		if coordMatch != None:
			unitMatch = unit.search(value, coordMatch.start(0))
		v = value
	return v
	
def repairStyle(node):
	num = 0
	if node.nodeType == 1 and len(node.getAttribute('style')) > 0 :	
		# get all style properties and stuff them into a dictionary
		styleMap = { }
		rawStyles = string.split(node.getAttribute('style'),';')
		for style in rawStyles:
			propval = string.split(style,':')
			if len(propval) == 2 :
				styleMap[propval[0].strip()] = propval[1].strip()

		# I've seen this enough to know that I need to correct it:
		# fill: url(#linearGradient4918) rgb(0, 0, 0);
		for prop in ['fill', 'stroke'] :
			if styleMap.has_key(prop) :
				chunk = styleMap[prop].split(') ')
				if len(chunk) == 2 and chunk[0][:5] == 'url(#' and chunk[1] == 'rgb(0, 0, 0)' :
					styleMap[prop] = chunk[0] + ')'
					num += 1

		# Here is where we can weed out unnecessary styles like:
		#  opacity:1
		if styleMap.has_key('opacity') :
			opacity = string.atof(styleMap['opacity'])
			# opacity='1.0' is useless, remove it
			if opacity == 1.0 :
				del styleMap['opacity']
				
			# if opacity='0' then all fill and stroke properties are useless, remove them
			elif opacity == 0.0 :
				for uselessStyle in ['fill', 'fill-opacity', 'fill-rule', 'stroke', 'stroke-linejoin',
					'stroke-opacity', 'stroke-miterlimit', 'stroke-linecap', 'stroke-dasharray',
					'stroke-dashoffset', 'stroke-opacity'] :
					del styleMap[uselessStyle]
					num += 1

		#  if stroke:none, then remove all stroke-related properties (stroke-width, etc)
		#  TODO: should also detect if the computed value of this element is fill="none"
		if styleMap.has_key('stroke') and styleMap['stroke'] == 'none' :
			for strokestyle in [ 'stroke-width', 'stroke-linejoin', 'stroke-miterlimit', 
					'stroke-linecap', 'stroke-dasharray', 'stroke-dashoffset', 'stroke-opacity'] :
				if styleMap.has_key(strokestyle) :
					del styleMap[strokestyle]
					num += 1

		#  if fill:none, then remove all fill-related properties (fill-rule, etc)
		#  TODO: should also detect if fill-opacity=0
		if styleMap.has_key('fill') and styleMap['fill'] == 'none' :
			for fillstyle in [ 'fill-rule', 'fill-opacity' ] :
				if styleMap.has_key(fillstyle) :
					del styleMap[fillstyle]
					num += 1
					
		#  stop-opacity: 1
		if styleMap.has_key('stop-opacity') :
			if string.atof(styleMap['stop-opacity']) == 1.0 :
				del styleMap['stop-opacity']
				num += 1
		
		#  fill-opacity: 1 or 0
		if styleMap.has_key('fill-opacity') :
			fillOpacity = string.atof(styleMap['fill-opacity'])
			#  TODO: This is actually a problem is the parent element does not have fill-opacity = 1
			if fillOpacity == 1.0 :
				del styleMap['fill-opacity']
				num += 1
			elif fillOpacity == 0.0 :
				for uselessFillStyle in [ 'fill', 'fill-rule' ] :
					del styleMap[uselessFillStyle]
					num += 1
		
		#  stroke-opacity: 1 or 0
		if styleMap.has_key('stroke-opacity') :
			strokeOpacity = string.atof(styleMap['stroke-opacity']) 
			#  TODO: This is actually a problem is the parent element does not have stroke-opacity = 1
			if strokeOpacity == 1.0 :
				del styleMap['stroke-opacity']
				num += 1
			elif strokeOpacity == 0.0 :
				for uselessStrokeStyle in [ 'stroke', 'stroke-width', 'stroke-linejoin', 'stroke-linecap', 
							'stroke-dasharray', 'stroke-dashoffset' ] :
					del styleMap[uselessStrokeStyle]
					num += 1

		# stroke-width: 0
		if styleMap.has_key('stroke-width') :
			strokeWidth = getSVGLength(styleMap['stroke-width']) 
			if strokeWidth == 0.0 :
				for uselessStrokeStyle in [ 'stroke', 'stroke-linejoin', 'stroke-linecap', 
							'stroke-dasharray', 'stroke-dashoffset', 'stroke-opacity' ] :
					del styleMap[uselessStrokeStyle]
					num += 1
		
		#  TODO: what else?
		
		# now if any of the properties match known SVG attributes we prefer attributes 
		# over style so emit them and remove them from the style map
		for propName in styleMap.keys() :
			if propName in svgAttributes :
				node.setAttribute(propName, styleMap[propName])
				del styleMap[propName]

		# sew our style back together
		fixedStyle = ''
		for prop in styleMap.keys() :
			fixedStyle += prop + ':' + styleMap[prop] + ';'
			
		if fixedStyle != '' :
			node.setAttribute('style', fixedStyle)
		else:
			node.removeAttribute('style')
		
	for child in node.childNodes :
		num += repairStyle(child)
			
	return num

# does nothing at the moment but waste time
def cleanPath(element) :
	path = element.getAttribute('d')

# converts raster references to inline images
# NOTE: there are size limits to base64-encoding handling in browsers 
def embedRasters(element) :
	global numRastersEmbedded

	href = element.getAttributeNS(NS['XLINK'],'href')
	
	# if xlink:href is set, then grab the id
	if href != '' and len(href) > 1:
		# find if href value has filename ext		
		ext = os.path.splitext(os.path.basename(href))[1].lower()[1:]
				
		# look for 'png', 'jpg', and 'gif' extensions 
		if ext == 'png' or ext == 'jpg' or ext == 'gif':

			# check if href resolves to an existing file
			if os.path.isfile(href) == False :
				if href[:7] != 'http://' and os.path.isfile(href) == False :
						# if this is not an absolute path, set path relative
						# to script file based on input arg 
						href = os.path.join(os.path.dirname(args[1]), href)				
				
			rasterdata = ''
			# test if file exists locally
			if os.path.isfile(href) == True :
				# open raster file as raw binary
				raster = open( href, "rb")
				rasterdata = raster.read()

			elif href[:7] == 'http://':
				# raster = open( href, "rb")
				webFile = urllib.urlopen( href )
				rasterdata = webFile.read()
				webFile.close()
			
			# ... should we remove all images which don't resolve?	
			if rasterdata != '' :
				# base64-encode raster
				b64eRaster = base64.b64encode( rasterdata )

				# set href attribute to base64-encoded equivalent
				if b64eRaster != '':
					# PNG and GIF both have MIME Type 'image/[ext]', but 
					# JPEG has MIME Type 'image/jpeg'
					if ext == 'jpg':
						ext = 'jpeg'

					element.setAttributeNS(NS['XLINK'], 'href', 'data:image/' + ext + ';base64,' + b64eRaster)
					numRastersEmbedded += 1
					del b64eRaster				

def properlySizeDoc(docElement):
	# get doc width and height
	w = SVGLength(docElement.getAttribute('width'))
	h = SVGLength(docElement.getAttribute('height'))

	# if width/height are not unitless or px then it is not ok to rewrite them into a viewBox	
	if ((w.units != Unit.NONE and w.units != Unit.PX) or
		(w.units != Unit.NONE and w.units != Unit.PX)):
	    return

	# else we have a statically sized image and we should try to remedy that	

	# parse viewBox attribute
	vbSep = re.split("\\s*\\,?\\s*", docElement.getAttribute('viewBox'), 3)
	# if we have a valid viewBox we need to check it
	vbWidth,vbHeight = 0,0
	if len(vbSep) == 4:
		try:
			# if x or y are specified and non-zero then it is not ok to overwrite it
			vbX = string.atof(vbSep[0])
			vbY = string.atof(vbSep[1])
			if vbX != 0 or vbY != 0:
				return
				
			# if width or height are not equal to doc width/height then it is not ok to overwrite it
			vbWidth = string.atof(vbSep[2])
			vbHeight = string.atof(vbSep[3])
			if vbWidth != w.value or vbHeight != h.value:
				return
		# if the viewBox did not parse properly it is invalid and ok to overwrite it
		except ValueError:
			pass
	
	# at this point it's safe to set the viewBox and remove width/height
	docElement.setAttribute('viewBox', '0 0 %s %s' % (w.value, h.value))
	docElement.removeAttribute('width')
	docElement.removeAttribute('height')

# this is the main method
# input is a string representation of the input XML
# returns a string representation of the output XML
def scourString(in_string):
	global numAttrsRemoved
	global numStylePropsFixed
	global numElemsRemoved
	doc = xml.dom.minidom.parseString(in_string)

	# for whatever reason this does not always remove all inkscape/sodipodi attributes/elements
	# on the first pass, so we do it multiple times
	# does it have to do with removal of children affecting the childlist?
	while removeNamespacedElements( doc.documentElement, unwanted_ns ) > 0 :
		pass	
	while removeNamespacedAttributes( doc.documentElement, unwanted_ns ) > 0 :
		pass
	
	# remove the xmlns: declarations now
	xmlnsDeclsToRemove = []
	attrList = doc.documentElement.attributes
	for num in range(attrList.length) :
		if attrList.item(num).nodeValue in unwanted_ns :
			xmlnsDeclsToRemove.append(attrList.item(num).nodeName)

	for attr in xmlnsDeclsToRemove :
		doc.documentElement.removeAttribute(attr)
		numAttrsRemoved += 1

	bContinueLooping = True
	while bContinueLooping:
		identifiedElements = findElementsWithId(doc.documentElement, {})
		referencedIDs = findReferencedElements(doc.documentElement, {})
		bContinueLooping = ((removeUnreferencedIDs(referencedIDs, identifiedElements) + vacuumDefs(doc)) > 0)

	# repair style (remove unnecessary style properties and change them into XML attributes)
	numStylePropsFixed = repairStyle(doc.documentElement)

	# remove empty defs, metadata, g
	# NOTE: these elements will be removed even if they have (invalid) text nodes
	elemsToRemove = []
	for tag in ['defs', 'metadata', 'g'] :
		for elem in doc.documentElement.getElementsByTagNameNS(NS['SVG'], tag) :
			removeElem = not elem.hasChildNodes()
			if removeElem == False :
				for child in elem.childNodes :
					if child.nodeType in [1, 3, 4, 8] :
						break
				else:
					removeElem = True
			if removeElem :
				elem.parentNode.removeChild(elem)
				numElemsRemoved += 1

	# remove unreferenced gradients/patterns outside of defs
	while removeUnreferencedElements(doc) > 0:
		pass
	
	while removeNestedGroups(doc.documentElement) > 0:
		pass

	while removeDuplicateGradientStops(doc) > 0:
		pass
	
	# clean path data
	for elem in doc.documentElement.getElementsByTagNameNS(NS['SVG'], 'path') :
		cleanPath(elem)

	# convert rasters references to base64-encoded strings 
	for elem in doc.documentElement.getElementsByTagNameNS(NS['SVG'], 'image') :
		embedRasters(elem)		

	# properly size the SVG document (ideally width/height should be 100% with a viewBox)
	properlySizeDoc(doc.documentElement)

	# output the document as a pretty string with a single space for indent
	out_string = doc.documentElement.toprettyxml(' ')
	
	# now strip out empty lines
	lines = []
	# Get rid of empty lines
	for line in out_string.splitlines(True):
		if line.strip():
			lines.append(line)

	# return the string stripped of empty lines
	return "".join(lines)

# used mostly by unit tests
# input is a filename
# returns the minidom doc representation of the SVG
def scourXmlFile(filename):
	in_string = open(filename).read()
	out_string = scourString(in_string)
	return xml.dom.minidom.parseString(out_string)

if __name__ == '__main__':

	# parse command-line arguments
	args = sys.argv[1:]

	# by default the input and output are the standard streams
	input = sys.stdin
	output = sys.stdout

	# if -i or -o is supplied, switch the stream to the file
	if len(args) == 2:
		if args[0] == '-i' :
			input = open(args[1], 'r')
		elif args[0] == '-o' :
			output = open(args[1], 'w')
		else:
			printSyntaxAndQuit()

	# if both -o and -o are supplied, switch streams to the files
	elif len(args) == 4 :
		if args[0] == '-i' and args[2] == '-o' :
			input = open(args[1], 'r')
			output = open(args[3], 'w')
		elif args[0] == '-o' and args[2] == 'i' :
			output = open(args[1], 'w')
			input = open(args[3], 'r')
		else:
			printSyntaxAndQuit()

	# else invalid syntax
	elif len(args) != 0 :
		printSyntaxAndQuit()

	# if we are not sending to stdout, then print out app information
	bOutputReport = False
	if output != sys.stdout :
		bOutputReport = True
		printHeader()

	# do the work
	in_string = input.read()
	out_string = scourString(in_string)
	output.write(out_string)

	# Close input and output files
	input.close()
	output.close()

	# output some statistics if we are not using stdout
	if bOutputReport :
		print " Number of unreferenced id attributes removed:", numIDsRemoved 
		print " Number of elements removed:", numElemsRemoved
		print " Number of attributes removed:", numAttrsRemoved
		print " Number of style properties fixed:", numStylePropsFixed
		print " Number of raster images embedded inline:", numRastersEmbedded
		oldsize = os.path.getsize(input.name)
		newsize = os.path.getsize(output.name)
		#sizediff = (min(oldsize, newsize)  / max(oldsize, newsize)) * 100;
		sizediff = (newsize / oldsize);
		print " Original file size:", oldsize, "kb; new file size:", newsize, "kb (" + str(sizediff)[:5] + "x)"